Qwen

Qwen3-32B

Qwen3-32B is an open-weights 32B-parameter model by Qwen. It is listed as in-progress on the local.ai leaderboard.

open weights
Parameters
32B
Benchmarks (local.ai snapshot)
τ²-bench
GAIA
GDPval
HuggingFace metadata · updated 2025-07-26
repo
Qwen/Qwen3-32B
license
apache-2.0
downloads
7,940,971
likes
732

Metadata refreshed from the HuggingFace API (snapshot 2026-08-15T15:53:42.405Z). Context length and long-form descriptions are the next pipeline stage.

Does it fit your hardware?

Detect or select your GPU first. Green = weights + context headroom, yellow = tight, grey = does not fit.

Q4_K_M 18.6GQ6_K 28.8GQ8_0 35.2GFP16 64G
Quantization≈ File sizeNote
IQ2_XXS8.6 GBExtreme compression, largest quality loss
IQ3_XXS11.5 GBVery small
Q3_K_M14.7 GBSmall
NVFP416 GBNVIDIA 4-bit
Q4_K_S17.3 GBCommon
Q4_K_M18.6 GBMost common sweet spot
Q5_K_M24.3 GBBalanced
Q6_K28.8 GBNear-lossless
Q8_035.2 GBNear-lossless
FP1664 GBOriginal weights

Actual files on HuggingFace

FileSize
model-00001-of-00017.safetensors3.69 GB
model-00004-of-00017.safetensors3.63 GB
model-00005-of-00017.safetensors3.63 GB
model-00006-of-00017.safetensors3.63 GB
model-00007-of-00017.safetensors3.63 GB
model-00008-of-00017.safetensors3.63 GB
GGUF: MaziyarPanahi/Qwen3-32B-GGUF · 270,421 downloads
FileSize
Qwen3-32B.Q2_K.gguf11.50 GBdownload
Qwen3-32B.Q3_K_L.gguf16.14 GBdownload
Qwen3-32B.Q3_K_M.gguf14.87 GBdownload
Qwen3-32B.Q4_K_M.gguf18.40 GBdownload
Qwen3-32B.Q5_K_M.gguf21.62 GBdownload
Qwen3-32B.Q6_K.gguf25.04 GBdownload
GGUF: Qwen/Qwen3-32B-GGUF · 70,141 downloads
FileSize
Qwen3-32B-Q4_K_M.gguf18.40 GBdownload
Qwen3-32B-Q5_0.gguf21.08 GBdownload
Qwen3-32B-Q5_K_M.gguf21.62 GBdownload
Qwen3-32B-Q6_K.gguf25.04 GBdownload
Qwen3-32B-Q8_0.gguf32.43 GBdownload

What you save self-hosting this model

Uses the model's own API price when it exists; otherwise the closest commercial equivalent.

Waiting for hardware detection…

RunLocal presents third-party leaderboard data (local.ai / Exo Labs, 2026-08) with its own fit and cost estimates. Verify pricing, licenses and file sizes at the source before deploying a model.